LinearOperators.jl

LinearOperators.jl copied to clipboard

Matrix(op) may generate incorrect result if op * x changes x

3

`Matrix(op)` may generate incorrect result in the case when `op * x` changes the values stored in` x`. Typical example is when `op` is a linear equation solver which overwrites...

andreasvarga

- [ ] limited-memory SR1 approximation - [X] in loop form, forward ~~and inverse~~: #14 - [ ] in compact form, forward ~~and inverse~~ - NB: inverse is trivial because...

dpo

Release 2.4.1

2

geoffroyleconte

Test for unbound parameters

1

https://docs.julialang.org/en/v1/stdlib/Test/#Test.detect_unbound_args cf: #241

dpo

Nosrati diagonal quasi Newton update

5

MohamedLaghdafHABIBOULLAH

`ERROR: LinearOperatorException("storage types cannot be promoted to a concrete type")`

6

``` using ADNLPModels, CUDA, LinearOperators, NLPModels nvar = 10 x0 = CuArray(rand(nvar)) nlp = ADNLPModel(sum, x0, gradient_backend = ADNLPModels.ReverseDiffADGradient) V = typeof(x0) xc = V(undef, nvar) Hs = V(undef, nvar)...

tmigot

Compressed LBFGS (forward) operator

25

@dpo, this is the first version of an implementation of a compressed LBFGS (forward) operator. I made a first `structure`, as well as a `Matrix` interface, and a `mul!` method....

paraynaud

`InverseLBFGSOperator` for GPU

5

This operator is used in JSOSolvers https://github.com/JuliaSmoothOptimizers/JSOSolvers.jl/blob/5c08163ee594d8f8d7e12a98b14f4ed7cd22424e/src/lbfgs.jl#L87C3-L87C63 but nothing allows GPU computation

tmigot

Separate 5-args constructor

5

geoffroyleconte

To remove the dependency on LDLFactorizations.

abelsiqueira